Biomacromolecules 2003, 4, 778-782
778
Uptake of Atmospheric Carbon Dioxide into Silk Fiber by Silkworms Jun Magoshi,*,†,§,# Toshihisa Tanaka,†,§,# Haruto Sasaki,‡ Masatoshi Kobayashi,§ Yoshiko Magoshi,†,§ Hidetoshi Tsuda,†,§ Mary A. Becker,¶ Shun-ichi Inoue,†,§ and Ken Ishimaru† Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Corporation (JST), Japan, National Institute of Agrobiological Sciences (NIAS), 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan, Faculty of Agriculture, Tokyo University, 1-1-1 Yayoi, Bunkyo, Tokyo 113-8657, Japan, and Faculty of Engineering, Fukui University, 3-9-1 Bunkyo, Fukui 910-8507, Japan Received January 7, 2003
The relation between the uptake of atmospheric CO2 and insect’s production of silk fiber has not yet been reported. Here, we provide the first quantitative demonstrations that four species of silkworms (Bombyx mori, Samia cynthia ricini, Antheraea pernyi, and Antheraea yamamai) and a silk-producing spider (Nephila claVata) incorporate atmospheric CO2 into their silk fibers. The abundance of 13C incorporated from the environment was determined by mass spectrometry and 13C NMR measurements. Atmospheric CO2 was incorporated into the silk fibers in the carbonyl groups of alanine, aspartic acid, serine, and glycine and the Cγ of aspartic acid. We show a simple model for the uptake of atmospheric CO2 by silkworms. These results will demonstrate that silkworm has incorporated atmospheric CO2 into silk fiber via the TCA cycle; however, the magnitude of uptake into the silk fibers is smaller than that consumed by the photosynthesis in trees and coral reefs. Introduction
The question that arises is from where does the excess C in the silk fiber come. We postulate that this carbon is derived from atmospheric CO2. Until now, the relationship between the uptake of atmospheric CO2 and insect’s production of silk fiber has not been investigated quantitatively. To confirm this assumption, we conducted 13CO2 treatments using four silkworm species (Bombyx mori, Samia cynthia ricini, Antheraea pernyi, and Antheraea yamamai) and a spider (Nephila claVata). The measurement of 13C incorporated from the environment was performed by mass spectrometry and 13C NMR measurements. Here, the δ13C value is the carbon isotope ratio in delta (δ13C ) ((13C isotopic abundance in the sample (atom %)/13C isotopic abundance in a standard substance (atom %)) - 1) × 1000). 13
Silkworms, as well as other insects such as bees, butterflies, and spiders, produce silk fibers. Silk fibers are mainly formed out of proteins synthesized within the cells of the silk gland using amino acids derived from their diet. The silk gland of silkworm, the largest organ of the 5th instar larva, secretes and stores the silk proteins. The weight of the silk gland increases rapidly within 7 days after initiation of 4th ecdysis. The major amino acids of silk protein are glycine, alanine, serine, and tyrosine. These amino acids represent about 90% of the total amino acid content of the silk fiber.1,2 The mechanism of silk protein biosynthesis is unique and includes a selective metabolic pathway. It is therefore intriguing from both biological and technological aspects to clarify the biosynthetic pathway of insects’ production. Under normal circumstances, the natural carbon isotopic ratio (δ13C) values in small animals such as insects and spiders are similar to those of their dietary profile.3 However, the δ13C values of silk fibers from silkworms reared on an artificial silkworm feed were noted to be substantially larger than that of the feed itself, having ratios of -22.32‰ ( 0.085‰ for the silk and -24.53‰ ( 0.087‰ for the feed. * To whom correspondence should be addressed. Telephone: +81-29838-7463. Fax: +81-29-838-7417. E-mail address:
[email protected]. † National Institute of Agrobiological Sciences (NIAS). # Jun Magoshi and Toshihisa Tanaka contributed equally to the work. § Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Corporation (JST). ‡ Tokyo University. ¶ Fukui University.
Materials and Experimental Methods Samples were obtained from four different silkworm species (B. mori, S. c. ricini, A. pernyi, and A. yamamai) and a spider (N. claVata). An isotope-labeled 13CO2 atmosphere was prepared from Ba13CO3 and 50% (w/v) H3PO4. Silkworms (5th instar larvae) subjected to the 13CO2 treatment were placed into containers (260 × 260 × 100 mm3) containing an isotope-labeled 13CO2 atmosphere and allowed to form cocoons in containers for a period of 3 days. Concentrations of CO2 in 13CO2 treatment containers were 0.2% (v/v) because the 13CO2 gas contained 1670 ppm. As a control experiment, additional silkworms were allowed to form cocoons in containers containing ambient air for 3 days. Concentrations of CO2 in the control containers were 0.03% (v/v) because the 13CO2 gas contained 3.63 ppm. All silkworms before cocooning, both treatments and controls,
10.1021/bm0340063 CCC: $25.00 © 2003 American Chemical Society Published on Web 03/11/2003
Uptake of Atmospheric CO2 into Silk Fiber
Figure 1. 13C isotopic abundance values (atom %) in silk fibers formed by B. mori, S. c. ricini, A. pernyi, A. yamamai, and a spider (N. clavata). The left bar of each column represents ambient air; the right bar of each column represents 13CO2-treated. Error bars represent standard error of the mean. The symbols ///, //, and / represent significant differences at the P < 0.001, P < 0.01, and P < 0.1 levels, respectively, in the t-test.
were fed an artificial food of known isotopic carbon composition. After cocooning, each cocoon shell was divided into the outer, middle, and inner layers from surface of cocoon (thickness ) 0.6-1.0 mm). Their δ13C values were measured with an elemental analyzer (NC2500, CE instruments, Milano, Italy) and a mass spectrometer (Delta Plus system, Finnigan MAT, Bremen, Germany). High-resolution solid-state 13C cross-polarization/magic angle spinning (CP/ MAS) NMR spectra were obtained using a Chemagnetics CMX-300 NMR spectrometer. The solution-state 13C NMR spectra were obtained using a JEOL (Japan Electron Optic Laboratory, Japan) JNM-LA400 NMR spectrometer. Results and Discussion For all of the cocoons formed under the 13CO2 treatment, the 13C isotopic abundance values in each layer were significantly higher than those in the cocoons produced in the control experiments. The 13C isotopic abundance values in the control cocoons were 1.087 atom %. Under the 13CO2 environment, the 13C values tended to be higher in the inner layer of the cocoons than in the outer layer. The abundance of 13C in the inner layer was 1.159 atom % showing an increase of 6.8% over that of the control. The percentage of crystalline regions in the cocoon filament is higher in the inner layer compared with the outer layer; such crystalline regions do not adsorb water.4 The results, as summarized in Figure 1, indicate that the difference in the 13C values of all layers in the shell of the cocoon after 13CO2 treatment was a consequence of the uptake of atmospheric CO2 during silk fiber formation and not due to the adsorption of CO2 contained in water vapor. To confirm the effect of elevated CO2 concentration in the container, a 12CO2 atmosphere was prepared from Ba12CO3 and 50% (w/v) H3PO4. The purity of Ba12CO3 is 99.99%, and the isotopic ratio (12C/13C) has a value of 94.2% to 5.8%. Concentrations of CO2 in the 12CO2 treatment containers were 0.2% (v/v), and 13CO2 concentration is 101 ppm. The 13C isotopic abundance values of the cocoon under
Biomacromolecules, Vol. 4, No. 3, 2003 779
Figure 2. 13C isotopic abundance values (atom %) in silk fibers formed by B. mori, S. c. ricini, A. pernyi, A. yamamai, and a spider (N. clavata). The left bar of each column represents ambient air; the right bar of each column represents 13CO2-treated. Error bars represent standard error of the mean. The symbols ///, //, and / represent significant differences at the P < 0.001, P < 0.01, and P < 0.1 levels, respectively, in the t-test.
the 12CO2 treatment and that of the control are same (1.0869 and 1.0867 atom %). These results show that the elevated CO2 concentration has no effect on the uptake of atmospheric CO2. Silk fibers actually consist of two types of protein, fibroin and sericin.5 A layer of sericin surrounds the two cores of fibroin in each filament. Weight ratio between fibroin and sericin is about 80:20. Sericin is known to have a higher solubility in water than fibroin. Consequently, most of the sericin is extracted from cocoon filaments by dissolution with hot water. Aqueous solutions containing sericin are freezedried to obtain a powder, and fibroin is the residual material after extraction. After the 13CO2 treatment, the 13C values of fibroin and sericin (1.140 and 1.137 atom %) extracted from the inner layer of B. mori cocoons were significantly higher than the values obtained in the controls (1.085 and 1.085 atom %) (P < 0.01, an increase of 5.1% and 4.8%). These results, shown in Figure 2, suggest that atmospheric CO2 is incorporated in both protein components of the cocoon filaments. The difference between the 13C values obtained in the 13CO2 treatment and controls was not due to the adsorption of CO2 contained in water vapor because both samples were thoroughly dried prior to analysis by mass spectrometry. To establish the mechanism for atmospheric 13CO2 uptake in silk fiber, 13C NMR experiments were conducted using the same samples as prepared for mass spectrometric analysis. Figure 3 shows 13C CP/MAS NMR spectra of sericin extracted from the inner layers of B. mori cocoons for both the control experiment and the one after treatment with 13CO2. On the basis of the reported amino acid composition of sericin,5 the protein is composed mainly of the amino acids serine (31.97%), aspartic acid (13.84%), and glycine (12.70%). The serine CR and Cβ carbon signals were clearly visible at 55.0 and 61.3 ppm. Furthermore, alanine Cβ and glycine CR carbon signals were observed along with a broad carbonyl carbon signal centered at 172.5 ppm. This broad peak includes several carbonyl carbons from aspartic acid, serine, and glycine. The 13C CP/MAS NMR spectra of the control and 13CO2-treated sericin were almost identical
780
Biomacromolecules, Vol. 4, No. 3, 2003
Figure 3. 13C CP/MAS NMR spectra of sericin from the inner layer of B. mori cocoons. High-resolution solid-state 13C CP/MAS NMR spectra were obtained using a NMR spectrometer operating at 75.57 MHz. The 1H 90° pulse width was 4.65 µs, and the contact time was 1.5 ms. The acquisition number was about 15 000 to achieve an appropriate S/N ratio, and the repetition time was 5.0 s. Samples were placed in a cylindrical rotor and spun at 3.5 kHz. A single asterisk represents a spinning sideband of the carbonyl peak.
with the exception of the relative signal intensities. The carbonyl carbon intensity of the 13CO2-treated sericin was 12% higher than that of the control. In general, the 13C CP/ MAS NMR signal intensity is dependent on the efficiency of cross-polarization. To quantify this difference, the dependency of the cross-relaxation time (TCH) and the spin-lattice H ) on the signal relaxation time in a rotating frame (T1F intensity was determined for the carbonyl carbon. The values H for TCH and T1F were 0.34 and 5.5 ms, respectively, for the 13 CO2 treated sample and 0.38 and 5.6 ms, respectively, for the control sample. The 13C CP/MAS NMR spectra of sericin were obtained at 1.5 ms contact time. These relaxation times indicate that the 12% signal intensity increase was significant. The NMR results correlate well with the mass spectrometry results, indicating that atmospheric 13CO2 was incorporated in the carbonyl carbon form in the silk fiber. Solution-state 13C NMR experiments were subsequently performed on samples hydrolyzed by hydrochloric acid to obtain high-resolution NMR spectra. The solution-state 13C NMR spectra of the protein hydrolysate of sericin obtained from the inner layer of B. mori cocoons after the 13CO2 treatment and from the control experiment are shown in Figure 4. In contrast to the solid-state 13C NMR spectra, all carbonyl carbon signals attributable to each different amino acid were clearly observed at around 170-172 ppm in the solution-state 13C NMR spectra. The solution-state 13C NMR signal intensity is typically modulated by the number of neighboring 1H due to the nuclear Overhauser effect (NOE)6 and molecular motion. The mobility of each amino acid is assumed to be equal because most peptide bonds in proteins are decomposed by hydrolytic cleavage. To quantify the signal intensity difference, the relative intensities of the Cd O 13C signals were determined by using the intensities of tyrosine Cζ, and alanine Cβ as reference standards. The 13C signal intensities of alanine CdO, aspartic acid CdO, serine CdO, glycine CdO, and aspartic acid Cγ in sericin obtained from the 13CO2-treated sample were found to be higher than those of the control sample with an increase in the relative
Magoshi et al.
Figure 4. 13C NMR spectra of sericin from the inner layer of B. mori cocoons. Protein was hydrolyzed by 6 N hydrochloric acid for 24 h at 110 °C without additives. The hydrochloric acid after hydrolysis was removed by drying under reduced pressure. The solution-state 13C NMR spectra were obtained at 25 °C using a NMR spectrometer operating at 100.40 MHz. The 1H nuclei were decoupled during the sampling. The 1H 45° pulse width was 10.80 µs. The acquisition number was about 12 000 to achieve an appropriate S/N ratio. The repetition time was 4.0 s. Samples were spun at 12 Hz. Sampling point number was 32 K. Solvent used was deuterated dimethyl sulfoxide, d6-DMSO.
Figure 5. 13C NMR spectra of fibroin from the inner layer of B. mori cocoons. Protein was hydrolyzed by 6 N hydrochloric acid for 24 h at 110 °C without additives. The hydrochloric acid after hydrolysis was removed by drying under reduced pressure. The solution-state 13C NMR spectra were obtained at 25 °C using a NMR spectrometer operating at 100.40 MHz. The 1H nuclei were decoupled during the sampling. The 1H 45° pulse width was 10.80 µs. The acquisition number was about 12 000 to achieve an appropriate S/N ratio. The repetition time was 4.0 s. Samples were spun at 12 Hz. Sampling point number was 32 K. Solvent used was deuterated dimethyl sulfoxide, d6-DMSO.
intensities by 16.6% ( 0.2%, 12.3% ( 0.2%, 14.5% ( 0.2%, 19.8% ( 0.2%, and 14.6% ( 0.2%, respectively. The relative intensities for each amino acid (alanine, aspartic acid, serine, glycine) in fibroin was also found to be increased (18.8% ( 0.9%, 20.1% ( 0.9%, 14.4% ( 0.8%, 12.9% ( 0.8%, and 15.7% ( 0.8%), shown in Figure 5. These increases in the intensity were significant and indicated that the increased 13 C for each amino acid was a consequence of the uptake of atmospheric 13CO2. These results confirm that atmospheric CO2 is incorporated in silk fiber directly by the silkworm. It has been reported that one metabolic pathway of the silkworms B. mori and S. c. ricini is the tricarboxylic acid (TCA) cycle.7,8 The solution-state NMR results present a model for the uptake of atmospheric CO2 by the TCA cycle
Uptake of Atmospheric CO2 into Silk Fiber
Figure 6. Metabolic pathway and uptake of atmospheric CO2 into silk protein in B. mori. The carbon atom of oxaloacetate CdO (C-1) is derived from atmospheric CO2. The carbon atoms of R-ketoglutarate CdO (C-1 or C-5) are derived from atmospheric CO2 by the tricarboxylic acid (TCA) cycle. Aspartic acid and glutamic acid, which contained atmospheric CO2, were also formed via the TCA cycle. The carbon of atmospheric CO2, which is incorporated into L-malate is converted to the carbon of serine CdO via the metabolic pathway of the TCA cycle in the silkworm. Glycine is formed from the interconversion of serine. Alanine is formed from pyruvate, which has incorporated atmospheric CO2.
in silkworms. There are numerous metabolic pathways for glucose, biosynthesis of amino acids, and CO2 incorporation (in a general way, reductive pentose phosphate cycle, reductive carboxylic acid cycle, and acetyl-CoA cycle); however, in this study, a simple mechanism is suggested on the basis of the current results. Atmospheric CO2 gas in the form of the hydrogen carbonate ion (HCO3-) enters the hemolymph system from the spiracles in the silkworms. The pathway from pyruvate to oxaloacetate by pyruvate carboxylase incorporates the transition of the hydrogen carbonate ion into oxaloacetate via the B. mori TCA cycle (Figure 6). Conversion of oxaloacetate to L-malate via the TCA cycle leads to integration of atmospheric CO2 into the amino acid structure. Therefore, the uptake of atmospheric CO2 into the amino acids (alanine, aspartic acid, serine, glycine) occurs via the TCA cycle. One explanation for the elevated CO2 incorporated in the inner layer is that the uptake mainly occurs during the silk production after the silkworms have stopped eating food and just prior to spinning. In this case, there would be a lack of glucose in their body and the metabolic pathway for uptake would be limited to the pathway from L-malate. This mechanism is a little different from the general metabolic pathway for CO2 uptake. Here, the uptake of atmospheric CO2 into serine and glycine occurs by the TCA cycle, the pathway used when the supply of
Biomacromolecules, Vol. 4, No. 3, 2003 781
glucose for biosynthesis is lacking. Thus, this lack of glucose causes a shift in the metabolic pathway to one utilizing more CO2. This is an anaplerotic reaction and important to the incorporation of atmospheric CO2 into silk fibers. Therefore, silkworms do not always incorporate CO2 into silk fibers. This present finding in the silkworm supports the metabolic pathway of the silkworms.7,8 A silk-spinning spider, N. claVata, was also subjected to the 13CO2 treatment. The 13C isotopic abundance value in the dragline silk from the web of spiders treated with 13CO2 was found to be 1.131 atom % (Figure 1). As observed in the case of the silkworms, the 13C values were higher compared with those of the control experiments. These results suggest that the uptake of atmospheric CO2 during the production of silk fiber may occur in many species of insects and arthropods producing silk fibers. The total increase in 13C integrated into the inner layer of the B. mori cocoon was 0.044 wt %, as calculated from the corresponding quantity of 13C in the control silk fibers and that of the 13CO2 treated fibers. The efficiency of CO2 uptake by silkworms and spiders was far lower than fixation occurring through photosynthesis by, for example, trees and coral reefs. However, the relationship between the uptake of atmospheric CO2 and silk fiber has not been reported previously. In this manuscript, the increased quantity of 13C in the silk fiber was determined quantitatively. Because silk contains 0.065% CO2 of its fresh weight, the amount of CO2 incorporated in silk fiber may be as much as 58.5 tons per year based on the world’s raw silk production by the silk industry (90 000 tons in 20019). Furthermore, many other insects and arachnid species, including moths, wasps, bees, butterflies, and spiders, also produce silk fibers. The overall quantity of CO2 incorporated by these organisms could be larger than expected. Recently, it has been shown that fibroin can be produced by the activity of yeast and bacteria or plants implanted with the appropriate genes.10-16 Accordingly, implanting genes into non-silk-producing insects or animals could lead to the biosynthesis of silk, or a valuable protein (silk or silklike), and CO2 utilization technology. Therefore, insects might be able to incorporate a fair amount of atmospheric CO2 in total. These findings illustrate the benefits of insects to our environment-conscious society. Acknowledgment. We thank Bernard Lotz of Institute Charles Sadron in France and Han Zhan of Akita Prefecture University in Japan for their suggestions. We also thank Masao Kato of National Institute of Agrobiological Sciences in Japan for his kindness in providing the silkworms. References and Notes (1) Fukuda, T.; Kirimura, J.; Matuda, M.; Suzuki, T. J. Biochem. 1955, 42, 341-346. (2) Kirimura, J. Bull. Seric. Exp. Stn. 1962, 17, 447-552. (3) Rundel, P. W., Ehleringer, J. R., Nagy, K. A., Eds. Stable Isotopes in Ecological Research; Springer-Verlag: New York, 1989. (4) Iizuka, E. Biorheology 1965, 3, 1-8. (5) Komatsu, K. In Polymeric Materials Encyclopedia; Salamone, J. C., Ed.; CRC Press: Boca Raton, FL, 1996; Vol. 10, pp 7711-7721. (6) Noggel, J. H.; Schirmer, R. E. The Nuclear OVerhauser Effect; Chemical Applications; Academic Press: New York, 1971. (7) Asakura, T.: Kawaguchi, Y.; Demura, M.; Osanai, M. Insect Biochem. 1988, 18, 531-538.
782
Biomacromolecules, Vol. 4, No. 3, 2003
(8) Asakura, T.; Baba, J.; Demura, M.; Osanai, M. Rep. Prog. Polym. Phys. Jpn. 1989, 32, 609-612. (9) Silk, raw and waste production (FAO Statistical Databases, http:// apps.fao.org/, 2001). FAO Production Yearbook 1999; Food and Agriculture Organization of the United Nation, Ed.; FAO: Rome, 2001; Vol. 53. (10) Prince, J. T.; McGrath, K. P.; DiGirolamo, C. M.; Kaplan, D. L. Biochemistry 1995, 34, 10879-10885. (11) Lewis, R. V.; Hinman, M.; Kothakota, S.; Fournier, M. J. Protein Expression Purif. 1996, 7, 400-406. (12) Fahnestock, S. R.; Irwin, S. L. Appl. Microbiol. Biotechnol. 1997, 47, 23-32.
Magoshi et al. (13) Fahnestock, S. R.; Bedzyk, L. A. Appl. Microbiol. Biotechnol. 1997, 47, 33-39. (14) Arcidiacono, S.; Mello, C.; Kaplan, D.; Cheley, S.; Bayley, H. Appl. Microbiol. Biotechnol. 1998, 49, 31-38. (15) Scheller, J.; Guhrs, K.-H.; Grosse, F.; Conrad, U. Nat. Biotechnol. 2001, 19, 573-577. (16) Lazaris, A.; Arcidiacono, S.; Huang, Y.; Zhou, J.-F.; Duguay, F.; Chretien, N.; Welsh, E. A.; Soares, J. W.; Karatzas, C. N. Science 2002, 295, 472-476.
BM0340063
